Neoplasm of the central nervous system (CNS) constitutes the largest group of solid tumors in children and are second only to leukemia and lymphoma in their overall frequency during childhood. The annual incidence is approximately 1500 to 200 with slight increase in the last decade. Medulloblastoma is the most common CNS tumor n children, with approximately 350 to 400 new cases diagnosed in the US each year. Because of the low incidence of this tumor, compared with the adult tumors such as colon or breast carcinoma, very limited basic research has been done to further the understanding of the biology of genetics of medulloblastoma. Despite multi-center clinical trials using increasingly sophisticated neurosurgical and other therapeutic interventions, the majority of the children with advanced medulloblastoma eventually die of progressive disease. Progress in treatment outcome has been hindered primarily by the lack of understanding of the biological properties and the prognostic indicators of these tumors. Recent advances in molecular biology have provided the tools to delineate the various steps of genetic alterations involved in the development of human cancer. Numerous proto-oncogenes and a few tumor suppressor genes have been identified and the specific roles of some of these genes in oncogenesis are becoming clear. In this project, I plan to define some of the basic biological properties and the genetic alterations in medulloblastoma. Specifically, this project will consist of four parts: 1. Establishing a brain tissue bank consisting of both tumor and matched normal tissues obtained at the time of surgery as discarded surgical materials. In addition, brain tissues will also be collected from patients undergoing lobotomy for nonmalignant disease or autopsy. 2. Establishing and characterizing medulloblastoma cell lines and normal brain cell cultures. In addition, normal brain cell cultures will also be immortalized to provide unlimited supply of normal cells for future studies. These tumor and normal cell lines will be characterized in terms of their growth parameters, tumorigenicity in nude mice, potential for differentiation into neuronal or glial cells, karyotype, as well as the effects of various cytokines, growth factors and chemotherapeutic agents on growth, differentiation and cytotoxicity. 3. Defining regions of chromosomal deletion in medulloblastoma by detailed allelotyping using PCR-based microsatellite polymorphic markers. In addition, loss of heterozygosity (LOH) involving known tumor suppressor genes will also be examined and positive results will be followed up by a search for mutations in the remaining allele using single strand conformational polymorphism and DNA sequencing. LOH patterns will be compared between high and low risk medulloblastomas as well as between primary and relapsed tumors to determine if these LOH markers correlate with progression of disease. 4. Identifying differentially expressed genes in medulloblastoma using arbitrarily primed PCR to generate differential RNA fingerprints from normal and medulloblastoma cell lines. Differentially expressed genes found in cell lines will be confirmed by similar studies using RNA from fresh tumors and their matched normal tissues. Focus will be placed on detecting tumor suppressor genes that are expressed in the normal tissues but not in the tumors. Candidates for novel tumor suppressor genes identified through this type of screening will be followed up by gene mapping and positional cloning.